cs.CV（2026-09-21）

📊 共 45 篇论文 | 🔗 9 篇有代码

🎯 兴趣领域导航

支柱二：RL算法与架构 (RL & Architecture) (20 🔗3) 支柱九：具身大模型 (Embodied Foundation Models) (12 🔗4) 支柱三：空间感知与语义 (Perception & Semantics) (5 🔗1) 支柱一：机器人控制 (Robot Control) (3) 支柱八：物理动画 (Physics-based Animation) (2) 支柱七：动作重定向 (Motion Retargeting) (2 🔗1) 支柱六：视频提取与匹配 (Video Extraction) (1)

🔬 支柱二：RL算法与架构 (RL & Architecture) (20 篇)

#题目一句话要点标签🔗⭐
1 CMAMBADEPTH: Self-supervised Monocular Depth Estimation with Channel Mamba and Hybrid Attention 提出CMambaDepth以解决单目深度估计中的信息交互瓶颈问题 Mamba depth estimation monocular depth
2 Classifier-Free Guidance in Flow Matching: Non-Autonomous Potentials, Overshoot, and Posterior-Mean Control 提出PMC-CFG以解决流匹配中的引导失真问题 flow matching classifier-free guidance
3 Toward a foundation model for forest point clouds 提出森林点云基础模型以解决森林属性提取问题 representation learning foundation model ✅
4 Relationally Grounded Latent World Models for Autonomous Driving 提出基于关系图的潜在世界模型以提升自动驾驶性能 world model world models representation learning
5 WorldCrafter: Consistent Video World Model with Implicit 3D-aware Memory 提出WorldCrafter以解决视频世界模型长时间一致性问题 world model world models distillation
6 Evaluating the Generalization of Neuroimaging Foundation Models on African Brain MRI 评估神经影像基础模型在非洲脑MRI上的泛化能力 JEPA foundation model
7 NeuIDO: Neural Intrinsic Dynamics Operator for Physics-Informed 4D World Models 提出NeuIDO以解决物理信息4D世界模型的动态建模问题 world model world models
8 AnalogDepth: Multi-view Geometry from FPV drones under Analog Video Transmission 提出AnalogDepth以解决FPV无人机模拟视频传输下的深度估计问题 distillation Depth Anything 3D reconstruction
9 ME-VLM:A Unified VLM for Embodied Cognition and Agent Coordination 提出ME-VLM以解决物理AI中的视觉语言理解问题 reinforcement learning distillation spatiotemporal ✅
10 SLICEChat: Progressive In-Encoder Token Pruning for Whole-Slide Pathology Language Models 提出SLICEChat以解决全幻灯片病理图像处理的可扩展性问题 Mamba large language model multimodal
11 Look Where It Counts: A Free, Label-Free Visual Evidence Signal for Fine-Grained Vision-Language Reasoning 提出一种无标签视觉证据信号以提升细粒度视觉语言推理 distillation large language model multimodal
12 MiTHras: Task-specific Hierarchical Semi-supervised Contrastive Masked Autoencoder for Mitotic Figure Analysis 提出MiTHras以解决有丝分裂图像分析中的自动化挑战 masked autoencoder contrastive learning
13 LiAuto-MindViT: A Hybrid Vision Backbone with Adaptive Bidirectional Mamba 提出LiAuto-MindViT以解决视觉理解中的方向性偏差问题 Mamba SSM
14 Positive Pair Geometry Matters: Optimal Transport for Contrastive Learning of Visual Representations 提出OTCLR框架以优化对比学习中的正样本几何结构 representation learning contrastive learning
15 VideoGen-Agent: Reinforcing Video Generation Agents 提出VideoGen-Agent以解决视频生成中的知识与一致性问题 reinforcement learning multimodal
16 PixelDiT2: Representation-Grounded Pixel Diffusion Transformers 提出PixelDiT2以解决像素空间扩散模型收敛慢的问题 representation learning foundation model
17 DTKDP: A Dual Teacher Knowledge Distillation and Pruning Framework for Lightweight Oriented SAR Ship Detection 提出DTKDP框架以解决轻量化SAR船舶检测问题 distillation
18 ChartJudgeBench: Evaluating LMM Judges for Chart-to-Code Generation 提出ChartJudgeBench以评估图表到代码生成中的LMM评判者 reinforcement learning multimodal
19 U-PEN Mamba: Progressive Expansion with Selective State-Space Modeling for Efficient Retinal Vessel Segmentation 提出U-PEN Mamba以解决视网膜血管分割中的挑战 Mamba ✅
20 HappyWorld-Bench 提出HappyWorld-Bench以评估世界模型的可靠性与一致性问题 world model world models

🔬 支柱九：具身大模型 (Embodied Foundation Models) (12 篇)

#题目一句话要点标签🔗⭐
21 DeCo: Efficient Decouple-to-Couple Learning for Multi-Task Visual Grounding 提出DeCo框架以解决多任务视觉定位中的特征冲突问题 large language model multimodal visual grounding ✅
22 Benchmarking Off-the-Shelf Multimodal AI Models Against Dermatologists on Patient-Captured Skin Images 评估多模态AI模型在皮肤病诊断中的应用效果 large language model multimodal
23 Preoperative Prediction of Microvascular Invasion in Hepatocellular Carcinoma by Integrating Multimodal Ultrasound and Clinical Data: A Multicenter Study 提出多模态超声与临床数据融合模型以预测肝细胞癌微血管侵犯 multimodal
24 Document Retrieval-Aware Chunking (D-RAC): Universal Retrieval-Aware Ingestion of Enterprise Documents via PDF Normalization and Multimodal Markdown Conversion 提出D-RAC以解决企业文档异构格式处理问题 multimodal
25 SPHQuant: Efficient extreme low bit weight quantization for Vision-Language Models 提出SPHQuant以解决极低比特宽度量化问题 foundation model multimodal ✅
26 Spatial Action Review: A Visual Analytics Dashboard for Auditing Language-to-Action Hand-offs in Electron Microscopy 提出Spatial Action Review以解决电子显微镜中的语言到动作交接问题 large language model multimodal
27 A$^2$Safe: Counterfactual Evidence-Aligned Adaptive Agent Collaboration for Safe and Effective Visual Question Answering 提出A$^2$Safe以解决视觉问答中的安全性与有效性问题 large language model multimodal
28 GAE: Learning a Geometry-Native Latent Space for 3D-Consistent World Generation 提出几何原生潜在空间以解决3D一致性生成问题 foundation model
29 INTCORT: Training-Free Spatial Reasoning Enhancement for Vision-Language Models via Input Transformations and Confidence Routing 提出INTCORT以解决视觉语言模型空间推理不足问题 multimodal
30 0.5\%>100\%: Bidirectional Reciprocal Learning for Referring Image Segmentation 提出双向互惠学习以解决图像分割中的语言视觉对齐问题 foundation model ✅
31 SRPR-Net: Semantic and Relational Prompt Refinement for Automated SAM-based Instance Segmentation 提出SRPR-Net以解决自动化实例分割中的语义引导不足问题 foundation model ✅
32 SAFe: Segment-guided Aggregation of Feature Densities for Anomaly-aware Segmentation 提出SAFe以解决视觉分割中的异常检测问题 foundation model

🔬 支柱三：空间感知与语义 (Perception & Semantics) (5 篇)

#题目一句话要点标签🔗⭐
33 Dynamic Thermal Gaussians: Multimodal 4D Gaussian Splatting 提出动态RGB-热成像重建框架以解决热传递时效性问题 gaussian splatting splatting spatiotemporal ✅
34 HyperCLIP++: Fine-tuning CLIP forOpen-vocabulary Semantic Segmentation in Hyperbolic Space 提出HyperCLIP++以解决开放词汇语义分割问题 open-vocabulary open vocabulary
35 InterHier: Learning Interconnected Hierarchical Semantics for Open-Vocabulary Object Detection 提出InterHier以解决开放词汇物体检测中的语义连接问题 open-vocabulary open vocabulary
36 Colon3R: Cross-Domain 3D Reconstruction from Monocular Colonoscopic Video 提出Colon3R以解决单目结肠镜视频的3D重建问题 3D reconstruction VGGT foundation model
37 When Wider Views Fail: Stress-Testing Feed-Forward 3D Reconstruction 提出应对视角变化的3D重建模型评估方法 3D reconstruction

🔬 支柱一：机器人控制 (Robot Control) (3 篇)

#题目一句话要点标签🔗⭐
38 An Unexpected Robot Policy: Early Evaluations of GPT-6 Astra on RoboDojo and Beyond 提出LLM作为机器人操控策略以提升任务成功率 manipulation bi-manual embodied AI
39 Dissecting Agentic Forensics: The Role of Triage, Prompting, and Evidence Arbitration in Open-World Fake Image Detection 提出基于代理AI的框架以解决开放世界假图像检测问题 manipulation large language model multimodal
40 Action-Slot: Structured Action-Centric Representation Learning for Multi-Agent Atomic Activity Understanding 提出Action-Slot框架以解决多智能体原子活动理解问题 sim-to-real representation learning

🔬 支柱八：物理动画 (Physics-based Animation) (2 篇)

#题目一句话要点标签🔗⭐
41 MECAIL: Communication-Aware Incremental Learning for Object Detection with 14.6 KB Spatiotemporal Experts 提出MECAIL以解决边缘设备增量学习的通信限制问题 spatiotemporal
42 High-resolution Nitrogen Dioxide Maps Reveal Exposure Limit Breaches across Europe 提出高分辨率NO2地图以解决欧洲空气质量监测不足问题 spatiotemporal

🔬 支柱七：动作重定向 (Motion Retargeting) (2 篇)

#题目一句话要点标签🔗⭐
43 Can Spiking Neural Networks play pinball? A neuromorphic motion detector for target tracking 提出一种全新脉冲神经网络以实现实时乒乓球游戏目标追踪 motion estimation
44 Do LiDAR Language Models Really Understand Spatio-temporal Relationships? 提出LiDAR-Hallu基准以评估时空关系理解能力 spatial relationship ✅

🔬 支柱六：视频提取与匹配 (Video Extraction) (1 篇)

#题目一句话要点标签🔗⭐
45 Evaluating Transformation Models for pCLE Mosaic Registration 提出新的变换模型以提高pCLE图像拼接精度 feature matching

⬅️ 返回 cs.CV 首页 · 🏠 返回主页